Sequence of multipolar transitions: Scenarios for URu2Si2

TL;DR

This paper classifies possible sequences of multipolar phase transitions, especially in URu2Si2, using symmetry analysis and mean field calculations, focusing on octupolar and quadrupolar orderings.

Contribution

It provides a group theoretical framework for understanding successive multipolar phase transitions, including scenarios involving octupole, quadrupole, and hexadecapole orders.

Findings

Classified possible second-order multipolar transitions in tetragonal systems.

Identified signatures of different multipolar orders in experimental measurements.

Suggested interpretations of URu2Si2 NQR results within the classification scheme.

Abstract

d- and f-shells support a large number of local degrees of freedom: dipoles, quadrupoles, octupoles, hexadecapoles, etc. Usually, the ordering of any multipole component leaves the system sufficiently symmetrical to allow a second symmetry breaking transition. Assuming that a second continuous phase transition occurs, we classify the possibilities. We construct the symmetry group of the first ordered phase, and then re-classify the order parameters in the new symmetry. While this is straightforward for dipole or quadrupole order, it is less familiar for octupole order. We give a group theoretical analysis, and some illustrative mean field calculations, for the hypothetical case when a second ordering transition modifies the primary T(xyz) octupolar ordering in a tetragonal system like URu2Si2. If quadrupoles appear in the second phase transition, they must be accompanied by a time-reversal-odd multipole as an induced order parameter. For O(xy), O(zx), or O(yz) quadrupoles, this would be one of the components of J, which should be easy either to check or to rule out. However, a pre-existing octupolar symmetry can also be broken by a transition to a new octupole--hexadecapole order, or by a combination of O(22) quadrupole and triakontadipole order. It is interesting to notice that if recent NQR results (Saitoh et al, 2005) on URu2Si2 are interpreted as a hint that the onset of octupolar hidden order at T0=17K is followed by quadrupolar ordering at T* = 13.5K, this sequence of events may fit several of the scenarios found in our general classification scheme. However, we have to await further evidence showing that the NQR anomalies at T* = 13.5K are associated with an equilibrium phase transition.